health

System Health Check

Run a comprehensive 10-subsystem health check across the TTS engine, Telegram bot, and supporting infrastructure. Produces a pass/fail report table with actionable fix recommendations.

Platform: macOS (Apple Silicon)

Self-Evolving Skill: This skill improves through use. If instructions are wrong, parameters drifted, or a workaround was needed — fix this file immediately, don't defer. Only update for real, reproducible issues.

When to Use This Skill

• Diagnose why TTS or Telegram bot is not working
• Verify system readiness after bootstrap or configuration changes
• Routine health check before a demo or presentation
• Investigate intermittent failures in the TTS pipeline
• Check for stale locks, zombie processes, or orphaned temp files

Requirements

• Bun runtime (for bot process)
• Python 3.14 with Kokoro venv at ~/.local/share/kokoro/.venv
• Telegram bot token in ~/.claude/.secrets/ccterrybot-telegram
• mise.toml configured in ~/.claude/automation/claude-telegram-sync/

Workflow Phases

Phase 1: Setup

Load environment variables from mise to ensure BOT_TOKEN and other secrets are available:

cd ~/.claude/automation/claude-telegram-sync && eval "$(mise env)"

Phase 2: Run All 10 Health Checks

Execute each check and collect results. Each check returns [OK] or [FAIL] with a brief diagnostic message.

Check 1: Bot Process

pgrep -la 'bun.*src/main.ts'

Pass if exactly one process is found. Fail if zero or more than one.

Check 2: Telegram API

BOT_TOKEN=$(cat ~/.claude/.secrets/ccterrybot-telegram)
curl -s "https://api.telegram.org/bot${BOT_TOKEN}/getMe" | jq .ok

Pass if response is true. Fail if false, null, or connection error.

Check 3: Kokoro venv

[[ -d ~/.local/share/kokoro/.venv ]]

Pass if the directory exists.

Check 4: MLX-Audio Import

~/.local/share/kokoro/.venv/bin/python -c "from mlx_audio.tts.utils import load_model; print('MLX OK')"

Pass if import succeeds with exit code 0.

Check 5: Apple Silicon

[[ "$(uname -m)" == "arm64" ]]

Pass if architecture is arm64. MLX-Audio requires Apple Silicon (M1+).

Check 6: Lock State

LOCK_FILE="/tmp/kokoro-tts.lock"
if [[ -f "$LOCK_FILE" ]]; then
  LOCK_PID=$(cat "$LOCK_FILE")
  LOCK_AGE=$(( $(date +%s) - $(stat -f %m "$LOCK_FILE") ))
  if kill -0 "$LOCK_PID" 2>/dev/null; then
    if [[ $LOCK_AGE -gt 30 ]]; then
      echo "STALE (PID $LOCK_PID alive but lock age ${LOCK_AGE}s > 30s threshold)"
    else
      echo "ACTIVE (PID $LOCK_PID, age ${LOCK_AGE}s)"
    fi
  else
    echo "ORPHANED (PID $LOCK_PID not running, age ${LOCK_AGE}s)"
  fi
else
  echo "NO LOCK (idle)"
fi

Pass if no lock or active lock with age under 30s. Fail if stale or orphaned.

Check 7: Audio Processes

pgrep -x afplay
pgrep -x say

Informational check. Reports count of running audio processes. Not a pass/fail -- just reports state.

Check 8: Secrets File

[[ -f ~/.claude/.secrets/ccterrybot-telegram ]]

Pass if the file exists and is non-empty.

Check 9: Stale WAV Files

find /tmp -maxdepth 1 -name "kokoro-tts-*.wav" -mmin +5 2>/dev/null

Pass if no stale WAV files found (older than 5 minutes). Fail if orphaned WAVs exist.

Check 10: Shell Symlinks

[[ -L ~/.local/bin/tts_kokoro.sh ]] && readlink ~/.local/bin/tts_kokoro.sh

Pass if symlink exists and points to a valid target within the plugin.

Phase 3: Report

Display results as a table:

| # | Subsystem        | Status | Detail                          |
|---|------------------|--------|---------------------------------|
| 1 | Bot Process      | [OK]   | PID 12345                       |
| 2 | Telegram API     | [OK]   | Bot @ccterrybot responding      |
| 3 | Kokoro venv      | [OK]   | ~/.local/share/kokoro/.venv     |
| 4 | MLX-Audio Import | [OK]   | mlx_audio module loaded         |
| 5 | Apple Silicon    | [OK]   | arm64 (MLX Metal)               |
| 6 | Lock State       | [OK]   | No lock (idle)                  |
| 7 | Audio Processes  | [OK]   | 0 afplay, 0 say                |
| 8 | Secrets File     | [OK]   | ccterrybot-telegram present     |
| 9 | Stale WAVs       | [OK]   | No orphaned files               |
|10 | Shell Symlinks   | [OK]   | tts_kokoro.sh -> plugin script  |

Phase 4: Summary and Recommendations

• Report total pass/fail counts (e.g., "9/10 checks passed")
• For each failure, recommend the appropriate fix or skill to invoke

TodoWrite Task Templates

1. [Setup] Load environment variables from mise in bot source directory
2. [Run] Execute all 10 health checks and collect results
3. [Report] Display results table with [OK]/[FAIL] status for each subsystem
4. [Summary] Show pass/fail counts (e.g., 9/10 passed)
5. [Recommend] Suggest fixes for any failures, referencing relevant skills

Post-Change Checklist

• All 10 checks executed (none skipped due to early exit)
• Results table displayed with consistent formatting
• Each failure has an actionable recommendation
• No sensitive values (tokens, secrets) exposed in output

Troubleshooting

Reference Documentation

• Health Checks - Detailed description of each check, failure meaning, and remediation
• Evolution Log - Change history for this skill

Post-Execution Reflection

After this skill completes, reflect before closing the task:

0. Locate yourself. — Find this SKILL.md's canonical path (Glob for this skill's name) before editing. All corrections target THIS file and its sibling references/ — never other documentation.
1. What failed? — Fix the instruction that caused it. If it could recur, add it as an anti-pattern.
2. What worked better than expected? — Promote it to recommended practice. Document why.
3. What drifted? — Any script, reference, or external dependency that no longer matches reality gets fixed now.
4. Log it. — Every change gets an evolution-log entry with trigger, fix, and evidence.

Do NOT defer. The next invocation inherits whatever you leave behind.